1 # Buildsheet autogenerated by ravenadm tool -- Do not edit.
7 SDESC[standard]= Implementation of Support Vector Machines (SVMs)
8 HOMEPAGE= http://svmlight.joachims.org/
11 DOWNLOAD_GROUPS= main patch
12 SITES[main]= http://osmot.cs.cornell.edu/svm_light/v6.02/
13 SITES[patch]= https://fastapi.metacpan.org/source/KWILLIAMS/Algorithm-SVMLight-0.09/
14 DISTFILE[1]= svm_light.tar.gz:main
15 DISTFILE[2]= SVMLight.patch:patch
16 DIST_SUBDIR= svmlight_602
18 SPKGS[standard]= single
20 OPTIONS_AVAILABLE= none
21 OPTIONS_STANDARD= none
27 LICENSE= CUSTOM1:single
28 LICENSE_NAME= CUSTOM1:"SVM-Light license"
29 LICENSE_FILE= CUSTOM1:{{WRKSRC}}/LICENSE.txt
32 FPC_EQUIVALENT= science/svmlight
37 MAKE_ARGS= CC="{{CC}}"
46 @${MKDIR} ${STAGEDIR}${PREFIX}/include/svmlight
47 ${INSTALL_DATA} ${WRKSRC}/*.h ${STAGEDIR}${PREFIX}/include/svmlight
48 .for file in svm_classify svm_learn
49 ${INSTALL_PROGRAM} ${WRKSRC}/${file} ${STAGEDIR}${PREFIX}/bin
51 ${INSTALL_DATA} ${WRKSRC}/libsvmlight.a ${STAGEDIR}${PREFIX}/lib/
52 ${INSTALL_LIB} ${WRKSRC}/libsvmlight.so \
53 ${STAGEDIR}${PREFIX}/lib/libsvmlight.so.1
54 ${LN} -s libsvmlight.so.1 ${STAGEDIR}${PREFIX}/lib/libsvmlight.so
56 [FILE:954:descriptions/desc.single]
57 SVMlight is an implementation of Vapnik's Support Vector Machine
58 [Vapnik, 1995] for the problem of pattern recognition, for the problem
59 of regression, and for the problem of learning a ranking function. The
60 optimization algorithms used in SVMlight are described in [Joachims,
61 2002a ]. [Joachims, 1999a]. The algorithm has scalable memory
62 requirements and can handle problems with many thousands of support
65 The software also provides methods for assessing the generalization
66 performance efficiently. It includes two efficient estimation methods
67 for both error rate and precision/recall. XiAlpha-estimates [Joachims,
68 2002a, Joachims, 2000b] can be computed at essentially no
69 computational expense, but they are conservatively biased. Almost
70 unbiased estimates provides leave-one-out testing. SVMlight exploits
71 that the results of most leave-one-outs (often more than 99%) are
72 predetermined and need not be computed [Joachims, 2002a].
76 aa48985a4c77eecd84d293de40d4731da767e49a1d2323c6198180652aa8724e 51026 svmlight_602/svm_light.tar.gz
77 0fc08fc2321970eb95e465c06deb7f99207ef90d7b492e3e6ff9883082a99962 19490 svmlight_602/SVMLight.patch
80 [FILE:139:manifests/plist.single]